(r.s.d.). Some of the analytes; i.e. hexachlorobenzene(HCB); heptachlor, and heptachlorexpoxide; 

 2,4' DDE; and 2,4' DDT; were present in concentrations near or below detection limits for one or 

 both Analytical Centers. The data for dieldrin and 2,4' DDD (Table 3) indicate between- laboratory 

 differences of a factor of two or three which has to be kept in mind when interpreting the field data. 

 MEL data are systematically slightly higher than GERG data when considering the entire set of 

 analytes (Figure 3.); but by less than a factor of two. Otherwise, the agreement between the two 

 laboratories for the Deer Island samples are within state-of-the-art limits for these types of 

 challenging analyses of trace concentration levels. 



B) Staten Island. 



Data from the Staten Island QA/QC intercomparison are presented in Tables 5 and 6 

 and Figure 4. The within-laboratory precision is between +/- 5 to 10% for those analytes with 

 reported concentrations well above the 250 pg/g dry weight detection limit; that is for 

 concentrations of 1 ng/g dry weight or above. There are between-laboratory differences of factors 

 of two to three for most of the chlorinated pesticides (Table 5). There is better agreement between 

 laboratories for several of the chlorinated biphenyl congeners, but there is a factor of two 

 difference for CB 52, CB153 and CB180. In contrast to the Deer Island QA/QC data, GERG 

 rather than MEL is systematically higher for the Staten Island samples (Figure 4). The main 

 difference between the Deer Island and the Staten Island QA/QC exercise was the state of the 

 samples when shipped to the laboratories. The Deer Island samples had been freeze dried whereas 

 the Staten Island samples were frozen wet samples. There may have been some difficulties in 

 determining wet weight to dry weight ratios which would account for systematic differences for all 

 analytes. 



C) NOAA-NIST. 



The NOAA-NIST sample results are presented in Tables 7 and 8, and Figure 5. There are 

 reasonable within laboratory precisions of the order of +/- 5 to 20% r.s.d. The between- 

 laboratory comparison indicates that, as with the Deer Island and Staten Island QA/QC samples, 

 there is a factor of two to three difference between the MEL and the GERG results for 2,4' DDD 

 and dieldrin with GERG reporting the higher concentration. There are also factors of four to five 

 difference between laboratories for the 4,4' DDE and 2,4' DDT concentrations. The 

 concentrations of 2,4' DDE, and heptachlor were near, at, or below detection limits for both 

 laboratories. The agreement between laboratories for individual chlorobiphenyl congeners shows 

 factors of three to ten differences for CBs 18, 28(31), 52, 44, 66/95, 101/90, 180 and 195; for 

 eight of the eighteen CBs analyzed. In contrast to the Deer Island results, the GERG data 

 appears to be systematically higher than the corresponding MEL data (Figure 5). 



18 



